> ## Documentation Index
> Fetch the complete documentation index at: https://ekacare-mintlify-b16ba99a.mintlify.site/llms.txt
> Use this file to discover all available pages before exploring further.

# Medication

## Overview

The **Medication** ontology links free-text drug mentions to Eka Care's
medication database — a curated catalogue of branded and generic drugs available
in India, with their compositions, forms and manufacturers.

| Property   | Value        |
| ---------- | ------------ |
| `ontology` | `medication` |
| `version`  | `v5.0`       |
| `metadata` | Optional     |

## When to use it

Use the medication ontology to code drug mentions from prescriptions, discharge
summaries or clinical notes — for example `Dolo 650` or `Paracetamol 500mg
Tablet`.

## How it works

The query is segmented into brand name, salt, form and volume; candidates are
retrieved from the medication database; and a hybrid matcher compares the
parsed query against each candidate's form, volume and salt.

A drug links (`is_linked: true`) only when the brand matches and every required
attribute — volume, form and qualifiers — matches exactly.

## The optional `metadata` field

`metadata` is not required, but it helps resolve ambiguous mentions:

| Key            | Description                                             |
| -------------- | ------------------------------------------------------- |
| `form`         | Medication form, e.g. `tablet`, `capsule`, `injection`. |
| `generic_name` | Generic / salt name of the drug.                        |
| `dose_unit`    | Dosage unit, e.g. `mg`, `mcg`, `ml`.                    |
| `dose_custom`  | Custom dosage string.                                   |

If the form and dose are already present in the query (e.g. `Dolo 650 Tablet`),
`metadata` adds little. It is most useful when the query is terse or the form is
implied.

## Reading the results

Medication results do not use a numeric `score` — read the `metadata` instead:

* `term_id` — the medication identifier (e.g. `b-4117370658`)
* `is_linked` — `true` when the drug is matched with all required attributes
* `metadata.generic_name`, `generic_composition`, `manufacturer_name`,
  `product_type`, `is_otc` — catalogue detail for the drug
* `metadata.matching_breakdown` — per-attribute match detail; when a drug does
  not link, `uncoded_reason` explains why (e.g. `volume_mismatch`)
* `metadata.linking_requirements` — attributes still needed to link confidently
  (e.g. `form_needed`)

The response also includes a `query_breakdown` showing the brand, form and
volume parsed from your request.

## Example

A query of `Dolo 650` with `metadata: { "form": "tablet", "generic_name":
"paracetamol", "dose_unit": "mg" }` links to `Dolo 650 Tablet`
(`b-4117370658`), composition Paracetamol (650mg), with `is_linked: true`.

## Try it out

See the [Link Entity API](/api-reference/health-ai/medical-entity-codification/link-entity)
and select the **Medication** example to try it.